172 results about "Discrete element model" patented technology

The invention discloses a discrete element-based three-dimensional fault / fracture network modeling method and device under a backlash pushing and covering system, and relates to the technical field of three-dimensional fault modeling, and the method comprises the steps: building an initialized fault plane model based on the seismic logging information parameters of a target region; extracting stratum parameters of the target area, and determining a deformation process in combination with an equilibrium profile recovery method; the method comprises the following steps: calibrating mesoscopic parameters of discrete elements according to stratum parameters, depositing a first discrete element model, optimizing and generating a second discrete element model, setting boundary conditions to carry out back-flushing pushing-covering simulation, comparing an output result of the second discrete element model with an actual geological section, adjusting model parameters until the output result and the actual geological section are matched, and establishing a three-dimensional fault and fracture network. And analyzing and establishing spatial distribution characteristics of the target area, and carrying out dynamic evaluation and trend prediction on the activity of the three-dimensional fault. According to the method, the geomechanical model is fused, the subjectivity of manual interpretation is overcome, and the problems of fault tip extension and the like in grid reconstruction are effectively avoided.

The invention discloses a dangerous rock falling motion feature analysis method based on discrete element simulation, and belongs to the technical field of engineering geology. Comprising the steps that dangerous rock data are collected through multiple means and preprocessed, and a three-dimensional visual dangerous rock mass initial model is constructed; based on the three-dimensional visual dangerous rock mass initial model, constructing a discrete element model, defining contact types and parameters among joint units by adopting a Hertz-mindlin contact theory, and determining normal contact force; based on the discrete element model and the normal contact force, a discrete element simulation technology is adopted, multi-physical field factors are considered to simulate dangerous rock falling movement, a discrete element model simulation result is obtained, a dynamic updating method is adopted for joint failure judgment, and the discrete element model is verified through an indoor physical model test and field monitoring data. And designing and implementing a prevention project according to a simulation result, thereby effectively preventing dangerous rock falling disasters and ensuring project safety.

The invention belongs to the technical field of discrete element simulation, and particularly relates to a digital twinborn discrete element model construction method based on CT (Computed Tomography) scanning, which comprises the following three steps of: in a CT scanning stage, acquiring a high-resolution cross-sectional image by optimizing scanning parameters, and carrying out image noise reduction and feature enhancement processing; in the structural analysis stage, a segmentation model is constructed based on a deep convolutional neural network, and automatic annotation of a scanned image and intelligent prediction of a three-dimensional space topological relation are realized; in the discrete element modeling stage, the reconstructed three-dimensional grid model is converted into a discrete element model with real physical attributes through parameter calibration, and discrete element simulation is achieved. According to the method, the non-destructive detection advantage of industrial CT and the feature extraction capacity of deep learning are combined, microstructure characterization is converted into macroscopic performance prediction, and a reliable simulation analysis method is provided for performance optimization of the composite material.

The invention discloses a discrete element contact parameter calibration device and method based on a dynamic repose angle, the device comprises a platform on which a motor and a motor control panel are fixedly arranged, the motor control panel is connected with a PC, the motor control panel is connected with the motor, and an output shaft of the motor is fixedly connected with a driving belt wheel; a first bearing seat and a second bearing seat are installed on the top face of the platform, the first bearing seat is rotationally connected with a first rotating shaft, the first rotating shaft is fixedly connected with a driven belt wheel connected with the driving belt wheel through a belt, and the two ends of the first rotating shaft are fixedly connected with driving wheels respectively. The second bearing seat is rotationally connected with a second rotating shaft; two ends of the second rotating shaft are fixedly connected with supporting wheels respectively; the cylinder is in contact with the surfaces of the driving wheel and the supporting wheel; a high-speed camera connected with a PC (Personal Computer) is fixedly arranged on the top surface of the platform, and discrete element analogue simulation software is installed in the PC. According to the invention, the dynamic angle of repose of the material in different flowing states can be measured in real time, and high-precision calibration of contact parameters among particles is facilitated through a discrete element model.

The invention provides a root-soil mixture impact penetration sounding dynamic response prediction method. The method comprises the following steps: firstly, constructing a discrete element model of a root-soil mixture, and constructing a root system structure with high simulation characteristics in the discrete element model; and fine simulation of interaction of the root system and the soil particles under the impact penetration power load is realized in combination with a contact model for simulating mechanical properties of the root-soil mixture, so that impact penetration penetration prediction of the root-soil mixture is finally realized. According to the method, various behaviors such as contact, friction, stripping and tearing between root-soil particles can be dynamically simulated on a microcosmic level, the behaviors are coupled into an integral model of impact penetration sounding, and the penetration resistance mechanism of a root-soil mixed medium can be accurately captured through the method, so that the penetration resistance of the root-soil mixed medium is accurately determined. And a reference is provided for abnormal identification and correction of the penetration signal.

The invention discloses a method for analyzing longitudinal force and longitudinal displacement of a long and large ramp continuously welded track under train braking, which comprises the following steps of: establishing an analysis / finite element model of the long and large ramp ballast track continuously welded track, taking initial ballast bed longitudinal resistance as an input parameter, and calculating to obtain initial distribution of longitudinal force and longitudinal displacement of a continuously welded track steel rail; meanwhile, the maximum value of longitudinal displacement of the continuously welded rail is obtained; and establishing a discrete element model of the long and large ramp ballast track seamless track, taking the maximum value of the initial distribution of the longitudinal displacement of the seamless track steel rail as an input condition, carrying out cyclic loading of the long and large ramp ballast track under the loading condition through a longitudinal displacement-longitudinal resistance dual-control method, extracting the longitudinal resistance of a track bed under the non-loading condition, and calculating the maximum value of the longitudinal displacement of the long and large ramp ballast track. And importing the data into the established analysis / finite element model to obtain updated distribution of longitudinal force and longitudinal displacement of the continuously welded rail. According to the method, the evolution characteristics of the longitudinal resistance of the long and large ramp ballast track seamless track ballast bed can be reasonably reflected.

The invention provides a shield hob type selection and tunneling parameter optimization method and device, a terminal and a medium, and belongs to the technical field of shield construction, and the method comprises the steps: obtaining the stratum information of a target project, the selectable type and / or value range of shield hob structure parameters, and the value range of tunneling parameters; constructing a rock mass discrete element model based on stratum information; presetting combinations of multiple groups of tunneling parameters and shield hob structure parameters; aiming at each group of combination, constructing a corresponding hob model, obtaining a rock breaking numerical simulation model based on the hob model and the rock mass discrete element model, inputting the tunneling parameters in the current combination into the rock breaking numerical simulation model for simulation calculation, and obtaining rock breaking specific energy and wear specific energy of the hob; and final shield hob structure parameters and tunneling parameters are determined. According to the method, the rock breaking numerical simulation model is constructed and the simulation calculation is performed, so that the rock breaking specific energy and the wear specific energy of the hob can be obtained, and the hob structure parameters and the tunneling parameters are cooperatively evaluated.

The invention discloses a flexible biological and interception model construction method based on particle contact and a simulation device, and the method comprises the steps: carrying out the improved contact model parameter determination based on various parameters of a marine organism according to a discrete element particle contact model of a target marine organism; and the flexible contact model is combined with the particle replacement model to generate a flexible target marine organism discrete element model, a target marine organism group is constructed, and a finite element / discrete element coupling calculation method is utilized to perform simulation evaluation on the interception efficiency of the interception net on the target marine organisms. The discrete element particle contact model combines the characteristics of efficient calculation of discrete elements and flexibility of biological materials, can accurately and quickly simulate and obtain the results of deformation, breakage, escape and the like of marine organisms under the action of the blocking net, and analyzes the interaction mechanism of the blocking net and the marine organisms according to the results. According to the method, the blank of describing the biological material by discrete elements can be filled, and the research on the marine organism intercepting mechanism of the intercepting net is facilitated.

The invention discloses a coal rock seepage-stress coupling simulation calculation method, which belongs to the technical field of geotechnical engineering numerical simulation, and comprises the following steps: carrying out gray threshold division on a coal rock sample CT scanning result to obtain a coal rock CT section; identifying particle position information in a coal rock CT section, and establishing a coal rock refined discrete element model; performing mechanical constitutive parameter calibration on the coal rock refined discrete element model; and on the basis of the coal rock refined discrete element model subjected to constitutive parameter calibration, establishing a fluid grid, setting model parameters, and carrying out seepage-stress coupling calculation. According to the method, the real form, size, spatial distribution and mutual inlaying relation of the mineral components in the coal rock can be accurately captured, and the most real basis is provided for accurate simulation of subsequent mechanical behaviors.

The invention discloses a thermal regeneration asphalt mixture functional reinforcement phase optimization design method based on discrete element simulation in the technical field of thermal regeneration road engineering material optimization, and the method comprises the steps: constructing a discrete element model of an asphalt mixture, extracting skeleton gap characteristic parameters, including the skeleton gap size, the volume ratio and the distribution condition, and obtaining a discrete element model of the asphalt mixture; according to the optimization design method, the mixing amount, the diameter range and the length parameters of the functional reinforcing material are optimized according to the characteristic parameters, the blindness of a traditional experiment depending on experience is avoided, and the design efficiency and performance of the functional asphalt mixture are improved. According to the method, the microstructure characteristics of the thermally recycled asphalt mixture are accurately quantified through the discrete element technology, the correlation model of the functional reinforced phase parameters and the skeleton gaps is established, the problems that a traditional test depends on experience and is low in efficiency are solved, and a high-performance material solution is provided for a functional asphalt pavement.

The invention discloses a loader load process prediction method based on MBD-DEM coupling simulation data driving, and is applied to the technical field of loader load prediction. Comprising the following steps: establishing a multi-rigid-body dynamic model of a loading machine working device; constructing a discrete element model of the material particles by adopting discrete element simulation software; mBD-DEM coupling is realized based on the multi-rigid-body dynamic model and the discrete element model; different working condition combinations are set, and load history data are obtained; preprocessing the load history data and the corresponding working condition data; constructing and training an LSTM neural network; and predicting the loading process of the loader based on the trained LSTM neural network. According to the method, the mechanical kinematics and dynamics characteristics of the loading machine working device and the discrete characteristics of materials can be considered at the same time, and the load prediction accuracy is improved.

The invention is suitable for the technical field of plant simulation modeling, and provides a corn ear general modeling method, which comprises the following steps: determining composition unit ball particles according to an ear structure; segmenting based on the shape and arrangement of the grains; obtaining morphological representation parameters and fitting a function relationship; geometric model parameters are calculated; universal modeling of the horse-tooth-shaped and fan-shaped grains is achieved; calculating the coordinates of spherical particles by adopting a combined ball method; filling the contour with spherical particles through rotation and translation transformation, and establishing a geometric model; and completing discrete element model construction in combination with mechanical model parameters. The method has the advantages of being high in universality, high in modeling precision, good in flexibility and the like, calculation and parameter calibration processes can be simplified, group model analysis is supported, and a reliable basis is provided for simulation of behaviors such as threshing and breaking of corn ears.

The invention is suitable for the technical field of plant modeling analysis, and provides a wheat leaf universal modeling method, which comprises the following steps: selecting a plurality of leaves of four wheat varieties, and testing and analyzing geometrical shape and size parameters and randomness and correlation thereof; solving the coordinates of the particles forming the balls by adopting a ball filling arrangement mode to obtain a geometric model of the stalks and four geometric models in which the balls formed by the three veins and the five veins are arranged at intervals and tangentially; measuring and calibrating contact mechanical parameters and adhesive mechanical parameters between the blades and between the blades and the boundary; adding the parameters into a mechanical model; and inputting the obtained geometric model and mechanical model of the leaf into discrete element software to obtain a discrete element model of the wheat leaf. The method provides important data support for modeling of the wheat leaves and other long-strip-shaped materials, and can be applied to simulation analysis of the contact action of the leaves and agricultural machinery parts in the wheat harvesting process and other technical fields needing wheat leaf modeling.

The application discloses a TBM cutter rock breaking efficiency optimization method and control system based on penetration and cutter spacing coupling, comprising: constructing a double-cutter rock breaking linear cutting numerical model, wherein the double-cutter rock breaking linear cutting numerical model is a discrete element model; simulating the double-cutter rock breaking linear cutting numerical model, and calculating the rock breaking specific energy consumption under the cutting action of different penetrations and cutter spacings according to the simulation result, wherein the rock breaking specific energy consumption is calculated according to mechanical parameters and lithology parameters; fitting the rock breaking specific energy consumption under different penetrations, and obtaining an optimal cutter spacing according to the fitting result, calculating the cutter spacing and penetration ratio according to the optimal cutter spacing; obtaining an optimal penetration and cutter spacing combination according to the trend of the cutter spacing and penetration ratio; and adjusting the penetration and cutter spacing of the TBM cutter according to the optimal penetration and cutter spacing combination, so as to improve the rock breaking efficiency and construction speed of the TBM.

The application discloses a soil deep loosening simulation method based on DEM-FEM bidirectional coupling. The method comprises the following steps: establishing a soil discrete element model and a deep loosening tool finite element model; using the discrete element DEM simulation to analyze the influence law of soil characteristic parameters and deep loosening tool motion parameters on soil deep loosening effect and deep loosening tool working resistance; changing the soil characteristic parameters and deep loosening tool motion parameters, and calculating the information such as soil deep loosening particle mixing degree, deep loosening tool working efficiency and working resistance under different variables; analyzing the influence law of the working resistance obtained from the DEM on the deep loosening tool structure through the finite element FEM analysis, changing the soil structure parameters and deep loosening tool structure parameters, and counting the statics and dynamics analysis results under different variables; simulating through the DEM-FEM bidirectional coupling, analyzing the soil deep loosening mechanism of the deep loosening tool, and obtaining the parameter and structure optimization results. Compared with the prior art, the application can effectively reduce resistance and consumption, and improve working efficiency.

The invention discloses an abrasive water jet rock breaking simulation method based on a DEM-CFD coupling method. Establishing an EBPM bonded rock mass discrete element model through a discrete element DEM, and calibrating mesoscopic parameters through a virtual mechanical test; simulating a water-abrasive two-phase flow through fluid dynamics CFD, and adopting a Gidaspow drag force model; the DEM-CFD bidirectional interface realizes real-time coupling; carrying out post-treatment to extract an erosion morphology and a crack network; the method can effectively simulate the rock mass fracture process and the large deformation behavior in the water jet rock breaking process and reveal rock mass damage evolution from the microscopic scale, and can be widely applied to research of water jet assisted rock breaking in the fields of civil engineering, mining, petroleum energy, marine resources and the like.

The application discloses a kind of discrete element modeling methods of multi-layer interpenetrating shear mixing drill and soil body, comprising: S1, the design parameter of multi-layer interpenetrating shear mixing drill is obtained, design parameter includes the pitch of blade, inclination, shape, size and quantity;Multi-layer interpenetrating shear mixing drill includes inner drill rod, multiple pairs of inner blades and outer frame arranged on inner drill rod;The outer frame includes multiple pairs of outer blades, two side plates and a collar arranged at intervals, and each pair of outer blades and each pair of inner blades are arranged at intervals;S2, the modeling of multi-layer interpenetrating shear mixing drill is carried out to obtain drill model;S3, establish soil model, import drill model, distribute contact model, and construct the coupling discrete element model between multi-layer interpenetrating shear mixing drill and soil particles;S4, simulation is carried out using coupling discrete element model, and quantitative analysis is carried out on mixing effect according to simulation result.The coupling discrete element model constructed by the above method has reliability and engineering applicability, and provides quantifiable basis for drill structure optimization.

The invention belongs to the technical field of lithium batteries, and relates to a material particle size grading optimization method and system based on a genetic algorithm and discrete element simulation. The method comprises the following steps: carrying out particle size distribution test on a material to be optimized to obtain a particle size distribution result as an optimizable particle size range; carrying out three-component peak separation on the obtained particle size distribution result, and approximately representing the particle size distribution of the material through three-component spherical particles; optimizing the stacking density of the three-component spherical particles in an optimizable particle size range by utilizing a genetic algorithm, and enabling the optimized stacking density of the three-component spherical particles to be approximate to the compaction density of the three-component spherical particles; obtaining physical property data of the material; and carrying out simulation verification on the compaction density of the three-component spherical particles obtained by optimization by using a discrete element model according to physical property data of a material. According to the method, optimization design is carried out on the particle size of the material based on discrete element analog simulation and a genetic optimization algorithm, and the large and small particle gradation of the high-compaction-density material can be rapidly screened.

The application relates to a kind of asphalt pavement road surface texture modeling methods based on three-dimensional discrete element model, comprising: step S1: selecting representative coarse aggregate from the aggregate constituting asphalt pavement wearing course, obtaining the surface topography as the first real aggregate profile model after scanning each selected coarse aggregate;Step S2: based on the first real aggregate profile model, the three-dimensional discrete element model of asphalt mixture is constructed as the asphalt pavement model, the micro-mechanical properties between the aggregate particles in the asphalt pavement model are given, and the texture parameters and the inter-particle porosity of the asphalt pavement model are calculated;Step S3: the surface texture parameters and the inter-particle porosity of the real asphalt pavement with the same aggregate and gradation are calculated, and compared with the texture parameters and the inter-particle porosity of the asphalt pavement model, if the error is too large, then return to step S2 to adjust the micro-mechanical properties between the aggregate particles in the asphalt pavement model.Compared with the prior art, the application can accurately simulate pavement texture and establish the corresponding relationship between wearing course texture and skeleton.

The invention belongs to the technical field of oil and gas field development engineering, and particularly relates to a conglomerate reservoir horizontal well perforation completion discrete element modeling method. The method comprises the following steps: extracting a conglomerate contour of a target area; taking the conglomerate contour of the extracted target area as a constraint boundary, generating a plurality of discrete small ball particles to fill the interior of the conglomerate contour and the residual matrix area, and further generating a conglomerate discrete element model; the method comprises the following steps: acquiring field perforation well completion data of a horizontal well in a target area, and converting the field perforation data to generate a horizontal well shaft wall unit containing perforations; the field perforation completion data at least comprise perforation depth, perforation aperture, perforation density, wellbore diameter and wellbore length parameters. Through the conglomerate reservoir perforation horizontal well fracturing simulation technology, the limitation of a traditional homogeneous model is broken through, and accurate reduction of the heterogeneity of a real reservoir is achieved.

The invention discloses a pavement load analysis method based on discrete elements and a related device, and the method comprises the steps: collecting geometric features and material parameters of an asphalt pavement, and constructing a discrete element model; based on the constructed discrete element model, creating a rigid load plate in the target area, and simulating a tire-road contact surface; a target load is applied through a rigid load plate, displacement field distribution under the action of the load is calculated based on a discrete element model subjected to the load, and settlement response analysis is achieved. The discrete element model of the asphalt pavement is established on a physical scale closer to reality, and interlayer and particle contact conditions are flexibly defined through fine unit division; on the premise of reducing the cost, the transmission path and the distribution characteristic of the dynamic load in an actual pavement structure and the complex internal mechanical response caused by the transmission path and the distribution characteristic can be more accurately captured, and powerful support is provided for comprehensive improvement of scientificity, reliability and precision of pavement engineering design and accurate prediction of pavement diseases.

The application discloses a kind of long approach in paste flow slope angle prediction method, belong to filling technical field, solve the prediction of filling slope angle in prior art is based on experience formula or static homogeneous model, unable to consider the influence of waste rock dynamic settlement on slurry flow resistance and slope form Problem, the method includes based on paste raw material basic parameter construction paste similarity model, based on range analysis combined with interaction analysis influencing factor on the influence degree of flow slope angle, convert flow slope angle calculation model into slope angle satisfaction function, obtain slope angle optimal filling parameter;The present application can comprehensively study the influence of multiple influencing factors and their interaction on paste flow slope angle through the pre-design and implementation of orthogonal experiment scheme, avoids the one-sidedness that single factor analysis may bring, makes the influence law obtained more comprehensive and in-depth, and is different from the homogeneous flow model or simplified discrete element model that the existing numerical simulation method is mostly used.

The invention relates to a numerical simulation method and device for tunnel mud burst and water burst disasters. The method comprises the following steps: acquiring a three-dimensional discrete element model of a target area according to geological information of the target area, simulating a mud burst water gushing disaster process caused by tunnel excavation by utilizing the three-dimensional discrete element model based on tunnel information, obtaining a numerical simulation result of the mud burst water gushing disaster, adjusting an angle corresponding to a tunnel crossing fault area in the tunnel information, and obtaining a numerical simulation result of the mud burst water gushing disaster. And obtaining numerical simulation results of the mud burst and water burst disasters under different tunnel information. By adopting the method, the numerical simulation result of the mud burst water gushing disaster simulated by the three-dimensional discrete element model can be obtained at the angles corresponding to different tunnel crossing fault areas, the occurrence process of the tunnel mud burst water gushing can be sufficiently described subsequently, and the accuracy of the simulation result is improved.

The invention discloses a three-dimensional random field-based rockfill material space variability discrete element simulation method. The method specifically comprises the following steps of: 1, constructing a triaxial test discrete element model and determining related parameters; step 2, generating a three-dimensional random field based on Karhunen-Loeve expansion; and step 3, random interpolation and assignment of particle contact parameters. Aiming at the technical defect that the real mechanical behavior of the rockfill material cannot be accurately reflected due to the fact that homogeneous hypothesis, two-dimensional modeling and continuous medium analysis are generally adopted in the prior art, the method comprises the following steps: constructing a three-dimensional random field of the rockfill material and introducing a discrete element particle system; more reasonable and more practical description of rockfill material space variability, macroscopic mechanical characteristics and mesoscopic mechanism deformation is realized.

The application discloses a modeling method of a two-dimensional ballast particle discrete element model and a modeling method of a two-dimensional sleeper-track bed-substrate discrete element model containing sand, and the modeling method of the two-dimensional ballast particle discrete element model comprises the following steps: extracting an outer contour of a ballast particle; determining an inner contour of the ballast particle according to the outer contour; filling the inner contour of the ballast particle with the largest inscribed circle disc until a preset standard is reached; filling a circle of inscribed circle discs between the inner contour and the outer contour of the ballast particle; exporting the radius and position coordinates of each disc after the filling is completed; and establishing a periodic boundary, and establishing a ballast particle discrete element model template according to the radius and coordinates of each disc. The two-dimensional ballast particle discrete element model provided by the application solves the contradiction between model precision and calculation cost, and the modeling method of the two-dimensional sleeper-track bed-substrate discrete element model containing sand can realize rapid filling of fine particles in a track bed ballast gap, and other dirty track beds can be quickly established from the model filled with fine particles once.

The present invention proposes a method for realizing virtual fatigue simulation of water-stabilized macadam materials based on a discrete element model. The method includes: (1) preparing indoor fatigue test specimens of water-stabilized macadam and conducting indoor fatigue tests; (2) constructing a fatigue damage evolution model of water-stabilized macadam materials based on the results of indoor fatigue tests; (3) using PFC software to establish a particle system of water-stabilized materials, setting contact models and contact parameters to form numerical specimens of water-stabilized macadam materials; (4) establishing a virtual fatigue decay model based on the fatigue damage evolution model, and using the Fish command to embed the virtual fatigue decay model into each running step of the fatigue test to simulate the decay behavior of the material properties of the model over time; (5) establishing a virtual fatigue loading system, applying fatigue cyclic loads to the numerical specimens of water-stabilized macadam materials described in step 3 to realize the virtual fatigue simulation of water-stabilized macadam materials.

The invention discloses a discrete element significant mesoscopic parameter screening method based on multi-model fusion. According to the method, based on a Plackett-Burman test, a Spearman correlation coefficient and a gradient boosting tree algorithm, the contribution degree, correlation and importance degree of a mesoscopic parameter relative to a response value are analyzed and calculated from the perspective of a linear regression model, a rank correlation coefficient and a decision tree, and the one-sidedness of a common method is made up; the Softmax function is adopted to quantify and fuse results of the three types of models, rationality and comprehensiveness of parameter screening are improved, complexity of subsequent parameter calibration tests can be effectively reduced, and powerful support is provided for rapid and accurate establishment of discrete element models.

A vehicle sand rolling simulation method, device, equipment and readable storage medium, comprising: constructing a whole vehicle simulation model and a discrete element model representing sand particles, and determining key contact areas and non-key contact areas in the whole vehicle simulation model; setting interaction parameters, including sand particle interaction parameters and interaction parameters between the discrete element model and the whole vehicle simulation model, wherein the contact relationship between the vehicle and the sand particles is defined by a discrete element and structure surface coupling mode for the key contact areas, and the acting force is calculated based on the geometric boundary of the sand particles; constructing a whole vehicle sand rolling simulation model according to the discrete element model, the whole vehicle simulation model and the interaction parameters; and performing vehicle sand rolling simulation based on the whole vehicle sand rolling simulation model. The whole vehicle is divided into key contact areas and non-key contact areas in the application, the contact relationship of the key areas is defined by a discrete element and structure surface coupling mode, the simulation fidelity is ensured, and the balance between calculation efficiency and simulation accuracy is realized.

The method for analyzing the formation and evolution characteristics of a core melt debris bed based on a particle method mainly comprises the following steps: 1, selecting a particle size to adapt to geometric modeling and coolant modeling; 2, selecting a melt material and a coolant material, setting initial parameters, and starting calculation; 3, calculating melt column fragmentation by using a moving particle semi-implicit method basic model; 4, calculating melt particle solidification and combination by using a melting and solidification model; 5, calculating melt debris collision and settlement to form a debris bed by using a discrete element model; 6, calculating debris bed cooling or remelting by using a decay heat model and a boiling heat transfer model; 7, judging whether the end time is reached, if not, advancing by a time step, and if yes, outputting a calculation result; and the method can predict the melt fragmentation form, debris size distribution, debris bed accumulation form, debris bed coolability and debris bed remelting risk probability after a core melt accident.

The invention relates to the technical field of rock mass stress field inversion, in particular to a stratified rock mass discrete element model crustal stress field inversion method based on deep learning, which comprises the following steps: acquiring a main control surface patch node extraction normal change sequence, comparing attitude path positioning direction mutation regions, analyzing included angle change to generate a construction path trend structural body, and calculating the stress field of the stratified rock mass. And node information is extracted to establish a structural path combination input primitive set, direction attributes are written to form a crack direction embedded tag group, and a constructed inversion sequence is input into a preset deep learning model to output a crustal stress inversion tensor field result. According to the method, the node direction change is extracted, mutation fragments are identified, spatial paths consistent in direction are constructed, a node sequence is generated in combination with the principal stress direction and the construction trend, coordinates, direction and construction information are fused to construct a spatial label group, an inversion input sequence is embedded, and the structural path and stress distribution relation is associated. The space matching between the path identification capability and the tensile stress direction is improved, and the distribution continuity and inversion stability of the tensor result in the structure are enhanced.